Alone atop the AFC at 11-3

[patsfaninpittsburgh]

The D is very bad in certain aspects:
- Total yds allowed: 414.4, 32nd in the NFL
- Pass yds allowed: 296.7, 32nd in the NFL
- Opp QB rating: 87.5, 24th in the NFL
- Opp completion %: 63.0, 26th in the NFL
- Yds per passing attempt allowed: 8.1, 29th in the NFL
(Lump all the passing stuff together and just say they have an awful pass defense)
- 3rd down conversions allowed: 44.5%, 29th in the NFL
- 1st downs allowed: 324, 32nd in the NFL
- defensive penalties: 104, 28th in the NFL
- Rush yds per attempt: 4.6, 26th in the NFL

The D is about average in certain aspects:
- Points allowed: 21.2, t14th in the NFL
- Rush yds allowed: 117.6, 19th in the NFL
- Pass td allowed: 21, t18th in the NFL
- Sacks: 33, 16th in the NFL
- Rush td allowed: 12, t19th in the NFL

The D is good or very good in certain aspects:
- Takeaways: 28, t1st in the AFC, t5th in the NFL
- Red Zone defense (TD %): 50.9%, 13th in the NFL

It's TOTALLY possible that the last two things - the very things NE is good at - are enough for this defense to actually play a positive role for the Pats in the postseason. They are very capable of allowing a team to march down the field on them, but they either stone them in the red zone and hold them to FG (while the Pats' offense gets TDs) or they cause a game-changing turnover.

If those things are not happening, this team will give up a lot of points.

Here is proof

Stats are for losers.

 

[ivanvamp]

Yeah, it was garbage time.

It's a matter of mind and matter.

I don't mind because in the scheme of things, it didn't matter.

You should become an offensive corrdinator. While other guys come up with the "Run 'n Shoot", "Wildcat", " Two TE Set"........................................

You can come up with the "Let's Get Blown Out For Three Quarters so We Can Have A Career Day 4th Quarter and Lose By Marginally Less" offense.

I bet the Dolphins, Eagles, Raiders, Jets, Colts would love your services.

Seriously, what on earth are you talking about. I'm going to repost this from what I said earlier b/c you must not have read it.
- - - - -
Henne threw for 416 yards. Of those, 101 came in "garbage time". At 5:55 of the 4th quarter, the score was NE 31, Mia 17, and the Dolphins had the ball on the NE 1 yard line. The previous play, Miami scored a TD but the call was reversed and the ball was placed on the 1. They were inches from cutting the lead to 1 touchdown with nearly 6 minutes left. So in just over 3 1/2 quarters, Henne threw for 315 yards.

Young threw for 400 yards. Garbage time began in that game at the 11:27 mark of the 4th quarter, when NE went up 31-13. From that point on, Young threw for 123 yards, meaning that for just over 3 quarters, Young threw for 277 yards. *VINCE YOUNG* threw for 277 yards. Guess how many 277+ passing yard games Young has had in his career. I'll tell you: 3. In 60 career games. So just twice in his career besides that game had he ever thrown for 277 yards.
- - - - -
Henne came within inches of making it a one TD game with just under 6 minutes left. That is absolutely, positively *NOT* garbage time. That game was very much in doubt at that point. Henne threw for 315 non-garbage time yards. That's a lot, especially for a guy like him, who is not very good. You are flat-out wrong to say that most of his yards came in garbage time.

Young had the third biggest passing day of his life before garbage time. Garbage time began at 11:27 of the 4th quarter, once NE went up by 18 points. Until then, Philly was down just 11 points with nearly a full quarter to go. That is NOT garbage time by any stretch of the imagination.

You couldn't be more wrong about these two particular situations. You're right about Orlovsky. I didn't look at any of the other cases. But you are 100%, completely, indisputably, wrong about Henne and Young - both had very big passing days against NE *before* garbage time.

:bricks:

 

[ivanvamp]

Here is proof

Stats are for losers.

So how do you evaluate the quality of the defense then? How good they look in uniform? How nice they are to underprivileged kids?

 

[lamafist]

Ahhhhhhh

Didn't like both teams win?

How do the stats compute in the first three quarters when both games were actually like.....being decided?

Also, how did that "elite" defense do against San Diego?

As I said, completely clueless.

You have zero fundamental ability to discern what matters to the outcome of a game vs what doesn't matter to the outcome of the game.

That's the poster child of clueless.

Or explain why Chad Henne, Vince Young, Dan Orlovsky, Mark Sanchez, Jason Campbell are waiting till the 4th quarter down multiple scores to start getting it done.

Are they calling for their inner Tebow?

Here are some interesting numbers...

Pats' points allowed by quarter: 1st-63 2nd-69 3rd-50 4th-115

The Packers have a similar pattern going: 1st=67 2nd=84 3rd=44 4th=112. In fact, of the seven teams that have given up 100+ 4th quarter points, every other team besides NE has at least one other quarter in which they've given up 84 points or more.

The Pats' allowed an average of 4.4 pts per quarter in the first three quarters. That's actually the exact same per-quarter average that Baltimore's D allowed in the first 3 quarters. The difference, in terms of scoring defense, is that the Ravens allowed only 53 4th quarter points all season, good for 2nd best in the league (behind Atlanta, who gave up 106 pts in the 3rd, go figure.)

Given that, among other factoring variables, the Pats' opponents had the worst average starting field position on their possessions, compared to the Ravens, who had the 22nd, I wouldn't go so far as to suggest that the Pats' defense was "as good as" the Ravens over the first three quarters.

I would, however, say that it seems like this defense is the exact opposite of a front runner -- it tends to fall apart when asked to simply play containment and slow the opposing team down.

 

[Ring 6]

That said, I think you can still very much discount the value of the Pats' points/allowed per game because of the fewer drives faced. If your team's style of play results in fewer drives per game, than you would naturally expect both your team and its opponents to score fewer than average points.

This doesn't make any sense to me.
How is points per drive a better metric than points per game when the style of defense leads to a different number of drives?

This is why per-drive stats are much more illuminating than per-game stats. Per-game stats contain an unknown variable that per-drive stats allows you to adjust for.

Like any stat, it lessens some biases and increases others.

Again, this is actually an argument against per-game stats and for per-drive stats. Per-game stats are hard to compare with one another because they are influenced by both team's style of play. Per drive stats are a simple metric that measures how effective your team was, on average, at stopping opponents from scoring during their possessions.

But those factors are also there to influence per drive stats.
If 50% of my drives are with a 14+ point lead, that is a variable that certainly will affect the results.

It doesn't matter if there were a lot of drives or a little, since teams alternate possessions, what you're able to do (or keep your opponent from doing) in the drives allotted is what counts.

I see your point, but again, you win or lose based upon the scoreboard, not a rendering of which statistical data is most likely to result in a win.
Statistical arguments are flawed from the start because they encompass incongruous situations, and are merely an attempt to predict what should have happened. The actual result doesn't need statistical analysis.

As I am sure you would agree, how the defense performs with a 17 point + lead in the 4th quarter is really irrelevant to winning or losing. The game is decided. If the Patriots were the worst ever in those situations (dismssing the garbage time argument) it would still be irrelevant to how they play when the game is undecided, unless they play the same way, which clearly they do not.
Gross statistics just aren't good predictors of results in a football game.

 

[Ring 6]

Here are some interesting numbers...

Pats' points allowed by quarter: 1st-63 2nd-69 3rd-50 4th-115

The Packers have a similar pattern going: 1st=67 2nd=84 3rd=44 4th=112. In fact, of the seven teams that have given up 100+ 4th quarter points, every other team besides NE has at least one other quarter in which they've given up 84 points or more.

The Pats' allowed an average of 4.4 pts per quarter in the first three quarters. That's actually the exact same per-quarter average that Baltimore's D allowed in the first 3 quarters. The difference, in terms of scoring defense, is that the Ravens allowed only 53 4th quarter points all season, good for 2nd best in the league (behind Atlanta, who gave up 106 pts in the 3rd, go figure.)

Given that, among other factoring variables, the Pats' opponents had the worst average starting field position on their possessions, compared to the Ravens, who had the 22nd, I wouldn't go so far as to suggest that the Pats' defense was "as good as" the Ravens over the first three quarters.

I would, however, say that it seems like this defense is the exact opposite of a front runner -- it tends to fall apart when asked to simply play containment and slow the opposing team down.

The next logical step would be to look at those 4th quarter points as see which were harmful (bills, giants for example) and which were just useless points that had no bearing on the game (Colts, Raiders for example)

 

[rojo8401]

Well, then sign me up for your "clueless" group, because it is quite obvious that the Pats defense has had troubles throughout the year and Henne, Orlovsky, VYoung, Grossman, etc. HAVE torched the defense (and not just in the second halves).

I'm often labeled a homer here and I have defended the defense consistently, believing that it was a work in progress. A week and a half ago I saw the game in person at FedEx and that defense was a shambles. The Skins were missing their only 2 outstanding offensive players - - Fred Davis and trent Williams. Roy Helu ran right over it. That was an OT game had Santana Moss not played shuffleboard at the end. The Broncos ran over it in the first half then inexplicably (perhaps because McGahee was out by then) went away from it. Give the Pats D and ST's credit in one very important area - - they DO create turnovers - - and that was the story for the Broncos game

All along I said 'wait 'til Chung and Spikes are back'. Well now Andre Carter is down, and we'll see how healthy Chung and Spikes are.

I still think in today's NFL this team could very well go all the way with the offense making up for the defense - - that's what Green Bay and New Orleans are doing. The great defenses (Pitt, Balto, etc.) conveniently have offenses and QB's that can very well be stopped (Flacco for lack of talent and Roethlisberger for lack of brains).

People on both sides are making silly over-generalizations to buttress their points.

The fact that the Pats have a subpar defense does not eliminate them from chances for a 4th SB. However, being willfully blind, at this point, to the fact that it is the team's weakness is your choice.

This D is not good. But the O is great enough to compensate. They have every bit of a chance as GB or NO has, and I think, the best chance in the AFC.

I await your wrath.

This is exactly how I see things. I don't always agree with you but you are spot on with this assessment.

 

[Ring 6]

So how do you evaluate the quality of the defense then? How good they look in uniform? How nice they are to underprivileged kids?

I think the absolute best way to do it is to assess whether the defense caused:
1) A win by doing their part
2) A win despite them
3) A loss that is their fault
4) A loss despite them playing well enough to deserve a win
Note that all of these should be considered up until the point the game was out of reach, meaning if you are up 38-7 in the 4th quarter, your defense contributed to the win, whether it ends up 56-7 or 38-28
This encompasses everything.
It takes into account game situation, clutch play, intangibles, and does not allow statistics accumulated due to circumstances to blur the answer.
A good example is to look at a QBs statistics.
It is very easy for a QB to build a big number of accumulated stats by throwing for a bunch of yards against a prevent defense when the game is over. But those yards were hollow, and the TD passes that came with them were as well. If the QB didn't play well enough to keep his team in the game, then amassed a bunch of yards, he looks statistically better than his opponent who had 3 good quarters then handed off all of the 4th quarter.
No statistic is good without persepctive.

 

[Ring 6]

The D is very bad in certain aspects:
- Total yds allowed: 414.4, 32nd in the NFL
- Pass yds allowed: 296.7, 32nd in the NFL
- Opp QB rating: 87.5, 24th in the NFL
- Opp completion %: 63.0, 26th in the NFL
- Yds per passing attempt allowed: 8.1, 29th in the NFL
(Lump all the passing stuff together and just say they have an awful pass defense)
- 3rd down conversions allowed: 44.5%, 29th in the NFL
- 1st downs allowed: 324, 32nd in the NFL
- defensive penalties: 104, 28th in the NFL
- Rush yds per attempt: 4.6, 26th in the NFL

The D is about average in certain aspects:
- Points allowed: 21.2, t14th in the NFL
- Rush yds allowed: 117.6, 19th in the NFL
- Pass td allowed: 21, t18th in the NFL
- Sacks: 33, 16th in the NFL
- Rush td allowed: 12, t19th in the NFL

The D is good or very good in certain aspects:
- Takeaways: 28, t1st in the AFC, t5th in the NFL
- Red Zone defense (TD %): 50.9%, 13th in the NFL

It's TOTALLY possible that the last two things - the very things NE is good at - are enough for this defense to actually play a positive role for the Pats in the postseason. They are very capable of allowing a team to march down the field on them, but they either stone them in the red zone and hold them to FG (while the Pats' offense gets TDs) or they cause a game-changing turnover.

If those things are not happening, this team will give up a lot of points.

You are assuming that if the good doesn't happen, the bad will.
There is a very convincing argument that the play of this defense is directly realated to the scoreboard. There for the negatives you are calling given may not be such a lock if the redzone D and takeaways were not keeping the score down.
There have been very, very few times where the defense allowed a score when already behind.
I just don't agree that you can take cumulative statistics, distregard situation, then create a sitaution and apply the cumulative stats to them.

 

[Palm Beach Pats Fan]

The D is good or very good in certain aspects:
- Takeaways: 28, t1st in the AFC, t5th in the NFL
- Red Zone defense (TD %): 50.9%, 13th in the NFL

It's TOTALLY possible that the last two things - the very things NE is good at - are enough for this defense to actually play a positive role for the Pats in the postseason.

We're 101-4 in the Belichick era when we win the turnover battle. This shows that one single statistic (well, a composite of two stats, giveaways on O and takeways on D) can correlate with winning more than any other thing I know of (well, besides points scored!).

If the D gets takeaways and the O doesn't give them back, sure we can have continued success. 101-4 is quite a sample size, so it is something to believe in.

 

[ivanvamp]

You are assuming that if the good doesn't happen, the bad will.

I am assuming that the defense is, in general, what the statistics say that it is. 14 games is a large enough data pool to get a pretty good picture; interestingly, that picture matches what the vast majority of people say is the eyeball test. In other words, the numbers pretty much match what it "looks like" is going on.

The Pats do some things well on D and do some other things poorly. Pass defense is one of the things they do poorly. They've been able to overcome poor pass defense (and it has been pretty poor on basically a weekly basis) by making red zone stops and getting key turnovers. I am suggesting that, yes, if they don't make red zone stops or get key turnovers, they are quite likely to struggle mightily on D.

There is a very convincing argument that the play of this defense is directly realated to the scoreboard. There for the negatives you are calling given may not be such a lock if the redzone D and takeaways were not keeping the score down.
There have been very, very few times where the defense allowed a score when already behind.

?? So this is just one component of bad defense. Another component is when you take a lead, you give it back. Another is when you go up by 2 scores, you promptly allow another one. They tend to give up a LOT of yards, which not only makes it harder to keep the other team off the scoreboard (it puts a lot of pressure on the D to make a red zone stop or get a turnover), it also puts more pressure on the offense. They are given fewer opportunities with the ball, and so they need to score on a higher percentage of offensive possessions. Moreover, because the D gives up a lot of yards, they are more likely to start with worse field position, thus making the offense do more work.

I just don't agree that you can take cumulative statistics, distregard situation, then create a sitaution and apply the cumulative stats to them.

I don't disagree. Tell you what: I've done a fair amount of research in my contributions to this thread. Why don't you come up with contextualized defensive performance. Break the Pats' D down by situation - opposing drives (yards, points, and time of possession) when the team is:

- down by >7
- down by 1-7
- tied
- up by 1-7
- up by >7

That will help make your argument.

 

[supafly]

I was at the Oakland game.

During the game, while most likely, patsfans.com was most likely in meltdown due to Jason Campbell "torching" the defense.......

The Raiders fans in my section kept loudly complaing that $*&^%@# Belichick was turning the game into a Brady vs Campbell game because McFadden wasn't a factor and that Campbell would screw up.

Imagine that, McFadden wasn't a factor and Campbell screwed it up.

That game for Oakland- highest yards and a "career day" and lowest points along with a blowout loss.

It's one thing to not be able to reconcile...it's another to not want to reconcile.....because the defense sucks.....because let's do what the Jets did the week before with an elite defense and get blown out while giving up 30+ points.

I totally agree with your statement about Belichick successfully taking away the biggest weapon, despite allowing major yardage. IIRC--that was one of Belichick's main positives in the Bills/Giants SB when he was the DC of the Giants. The plan was to allow Thurman Thomas to chew up clock and limit their own possessions to some degree, all the while knowing that he most likely would end up with a ton of rushing yds.

As we all know, the plan worked, and the NYGiants and BB won the Super Bowl.

It goes without saying that we are increasing our chances dramatically by taking away the biggest weapon/weapons from the other team. A lot of the time that comes with giving up a lot of yards.

Personally, I couldn't care less about giving up yards, especially when the standings point to a good shot at a deep playoff run once again. The problem that I have (and a lot of other posters) is that the current version of the defense just doesn't seem to be very good. The ability of some pretty average QB's to run up and down the field against our D is what scares a lot of people.

While good posters like shmessy point out that the offense is good enough to make up for the difference, the ability to march up and down the field at will is what bothers and worries some of us.

Whether that aspect of the defense will hinder us come playoff time remains to be seen, but that is a good point that you make when comparing 'overall' success (W vs. L), taking away the best player etc--even though it gives up a lot of yards. I think we all know BB couldn't care one way or another about giving up yards, as long as the scores remain lower and the team wins (while hopefully showing improvement).

We'll find out soon enough if the defense hindered the teams' chance or not.

 

[shmessy]

Sorry to step in here again, but I already debunked this myth when it comes to Henne and Young.

Henne.........Young.......man!

Take my wife.....please!

 

[Ring 6]

I am assuming that the defense is, in general, what the statistics say that it is.

Your comment was if they stop doing well in the red zone or getting turnovers they will allow a lot of points. That is NOT assuming it is what it is, because you are eliminating what they do well.

14 games is a large enough data pool to get a pretty good picture; interestingly, that picture matches what the vast majority of people say is the eyeball test. In other words, the numbers pretty much match what it "looks like" is going on.

Once again, you are talking about what will go on if they stop doing what they do well and continue doing what they do poorly. That answer is not rocket science and really is just a fantasy.

The Pats do some things well on D and do some other things poorly. Pass defense is one of the things they do poorly. They've been able to overcome poor pass defense (and it has been pretty poor on basically a weekly basis) by making red zone stops and getting key turnovers. I am suggesting that, yes, if they don't make red zone stops or get key turnovers, they are quite likely to struggle mightily on D.

If Aaron Rodgers stops throwing accurately the Packers won't score many points.

?? So this is just one component of bad defense. Another component is when you take a lead, you give it back. Another is when you go up by 2 scores, you promptly allow another one. They tend to give up a LOT of yards, which not only makes it harder to keep the other team off the scoreboard (it puts a lot of pressure on the D to make a red zone stop or get a turnover), it also puts more pressure on the offense. They are given fewer opportunities with the ball, and so they need to score on a higher percentage of offensive possessions. Moreover, because the D gives up a lot of yards, they are more likely to start with worse field position, thus making the offense do more work.

So you are using what they have done while winning and turning it into what will cause them to lose?
All of the implications of the statistics are part of the overall performance that is already in the book. I don't get how the argument is the way they allowed x number of points is the proof they aren't as good as x number of points.

I don't disagree. Tell you what: I've done a fair amount of research in my contributions to this thread. Why don't you come up with contextualized defensive performance. Break the Pats' D down by situation - opposing drives (yards, points, and time of possession) when the team is:

- down by >7
- down by 1-7
- tied
- up by 1-7
- up by >7

That will help make your argument.

If I was going to do that I already would have.
I am not real interested researching what seems to be pretty obvious to me. I do not need to convince myself of my own conclusion.
If you wish to base your opinion on accumulated stats without regard to the circumstances or game situations, that is fine by me, but I won't find much value in it.

 

[Joker]

Henne.........Young.......man!

Take my wife.....please!

I take my wife everywhere...


she finds her way home....rimshot

 

[ivanvamp]

Your comment was if they stop doing well in the red zone or getting turnovers they will allow a lot of points. That is NOT assuming it is what it is, because you are eliminating what they do well.

Once again, you are talking about what will go on if they stop doing what they do well and continue doing what they do poorly. That answer is not rocket science and really is just a fantasy.

I don't think this is difficult to understand, Andy. The Patriots' D gives up a fair amount of points and a TON of yards, especially passing yards. That's incontrovertible fact. One may wonder why the Pats are able to give up fewer points than their yards, 3rd down conversion rate, etc., all suggest they possibly should. I suggest that it's because of two things they do well: get turnovers and make red zone stops. And yes, I'm suggesting that if they don't do those things (one of which is certainly more unpredictable and unreliable (turnovers) on a game-to-game basis), they're likely to give up a lot of points.

If Aaron Rodgers stops throwing accurately the Packers won't score many points.

Correct. His accurate passing is the #1 reason Green Bay scores a lot of points and wins. If the other team can disrupt this, they stand a better chance of winning. Glad you're getting it.

If I was going to do that I already would have.
I am not real interested researching what seems to be pretty obvious to me. I do not need to convince myself of my own conclusion.

So let me get this straight. I present the statistical data to show what the Patriots do well, what they do kind of average, and what they do poorly. Another guy says that "stats are for losers". I then wonder how else one is supposed to measure the quality of a defense if not by the statistical measurements their performance produces. You then say that the stats are, essentially, meaningless, unless they are understood in their situational context.

I then say, ok, show me that this data is painting an inaccurate picture.

You then come back with: tough crap, do that research yourself?

Andy, that's not how it works. I have offered a viewpoint based on a myriad of statistics. I've also shown how guys like Vince Young and Chad Henne actually had huge passing days *before* garbage time, thus disproving another poster's claim to the contrary. See, how this works is that if you make a claim, the onus is on you to prove it somehow - either by argumentation or by statistical evidence (or some combination).

You are offering a claim - that the Pats' defensive statistics don't really tell the true story, that if the data were seen in their context it would show something different. It is up to YOU to support that claim. You can't just throw it out there and expect others to do your research for you.

You want to argue that point? Fine. Prove it.

If you wish to base your opinion on accumulated stats without regard to the circumstances or game situations, that is fine by me, but I won't find much value in it.

I don't give a crap if you find much value it in. If you're off in your own little world convinced that the data doesn't really show what it certainly appears to show, that's ok with me. But you entered the discussion challenging a point I was making. It appears that *you* are trying to convince *me* of something. I'm willing to be convinced. But I'm not going to do your job for you. You want me to be convinced? Convince me. Give me data. If you *aren't* trying to convince me, then why on earth did you bother jumping into the discussion to offer that "correction"?

 

[lamafist]

This doesn't make any sense to me.
How is points per drive a better metric than points per game when the style of defense leads to a different number of drives?

Like any stat, it lessens some biases and increases others.

But those factors are also there to influence per drive stats.
If 50% of my drives are with a 14+ point lead, that is a variable that certainly will affect the results.

I see your point, but again, you win or lose based upon the scoreboard, not a rendering of which statistical data is most likely to result in a win.
Statistical arguments are flawed from the start because they encompass incongruous situations, and are merely an attempt to predict what should have happened. The actual result doesn't need statistical analysis.

As I am sure you would agree, how the defense performs with a 17 point + lead in the 4th quarter is really irrelevant to winning or losing. The game is decided. If the Patriots were the worst ever in those situations (dismssing the garbage time argument) it would still be irrelevant to how they play when the game is undecided, unless they play the same way, which clearly they do not.
Gross statistics just aren't good predictors of results in a football game.

Most of the unaccounted-for variables you bring up would apply equally to per-drive and per-game stats, as well as to yardage and point stats. If you're looking for metrics that allow for these factors, I'd point you to FootballOutsider's DVOA et al., or to the stats kept at AdvancedNFLStats.com, particularly Win Probability based stats, as this metric is entirely based upon context-sensitive calculations of the net change in the likelihood of winning, as established by the results of the last 20 years of NFL football.

As for the per-drive v. per-game question:

Assuming a static level of offensive production, a style of defense that results in fewer possessions and might give up fewer points, but since it also results in fewer possessions for its offense, it needs to give up fewer points to result in the same number of wins.

Essentially, changing the number of possessions in the game creates an 'exchange rate' for points -- a touchdown in a 10 possession game is worth a lot more than a touchdown in a 14 possession game.

Per-game stats don't account for this. Per-drive stats avoid the issue by not grouping possessions by game.

Both stats are equally affected by things like garbage-time, level of competition, etc. Per-drive stats just have one fewer distorting variable.

 

[Ring 6]

I don't think this is difficult to understand, Andy.

Its not a matter of understanding your argument, that is easy. Its a matter of disagreeing with it.

The Patriots' D gives up a fair amount of points and a TON of yards, especially passing yards. That's incontrovertible fact.

As are takeaways and red zone defense. Your comment that I am discussing wants to make yards allowed an incontrovertible fact and red zone defense and takeaways subject to you pretending they don't exist or are somehow based on luck, mysterious forces or smoke and mirrors.

One may wonder why the Pats are able to give up fewer points than their yards, 3rd down conversion rate, etc., all suggest they possibly should. I suggest that it's because of two things they do well: get turnovers and make red zone stops. And yes, I'm suggesting that if they don't do those things (one of which is certainly more unpredictable and unreliable (turnovers) on a game-to-game basis), they're likely to give up a lot of points.

This is the flaw in your argument. You simply cannot form an honest argument by saying the variables that might support your conclusion are set in stone and the ones that don't will change.
There is certainly reason to believe that if the red zone defense and takeaway performance were not as good then the yardage statistics may be very different too, since all teams, and especially this one make game plan and play call decisions based upon situational football.


.If your argumnet is that if the defense never takes the ball away and stinks in the red zone it will allow a lot of points, that would hold fro every team in the NFL, so I'm not sure why you are making a point of it.

Correct. His accurate passing is the #1 reason Green Bay scores a lot of points and wins. If the other team can disrupt this, they stand a better chance of winning. Glad you're getting it.

So we are saying that if teams and players play well they win, and if they play poorly they lose? Not exactly earth-shattering or discussion worthy.

So let me get this straight. I present the statistical data to show what the Patriots do well, what they do kind of average, and what they do poorly. Another guy says that "stats are for losers". I then wonder how else one is supposed to measure the quality of a defense if not by the statistical measurements their performance produces. You then say that the stats are, essentially, meaningless, unless they are understood in their situational context.

That seems like a rudimentary understanding that wouldn't require debate.

I then say, ok, show me that this data is painting an inaccurate picture.

You then come back with: tough crap, do that research yourself?

I don't know where you get 'tough guy' out of me declining to accept an assignment you decided to give me.
I've seen every play of every game, I have a pretty good idea what that would show, so it is not worth my time to scour the internet to find that information.
If you feel it is needed, feel free.

Andy, that's not how it works. I have offered a viewpoint based on a myriad of statistics. I've also shown how guys like Vince Young and Chad Henne actually had huge passing days *before* garbage time, thus disproving another poster's claim to the contrary.

I'm not sure what the other posters claim was, I was discussing this with you.
This board has whined about the stats put up by 'bad QBs' while we win games for years. It means little to me. There is clearly a defensive philosophy on this team that they will take away a strength at the expense of allowing a weakness to succeed to a degree. We probably have the best record in the NFL vs bad QBs over the last 10 years by miles and miles, while probably also allowing more yards to them than most teams. I don't know why you don't want to recognize that, or want to give it more meaning.
Regardless of stats you are showing, there were a large amount of yards that WERE accumulated in 'garbage time'. I don't think anyone said they were all in garbage time. I don't think anyone said we have a great yards allowed stat outside of garbage.
People are simply, and correctly, saying that the cumulative defenses numbers are skewed by garbage time much more than other teams.

See, how this works is that if you make a claim, the onus is on you to prove it somehow - either by argumentation or by statistical evidence (or some combination).

I'm glad you have a model of 'how this works'. I don't care to follow your model. Dub yourself winner if thats what is important to you.

You are offering a claim - that the Pats' defensive statistics don't really tell the true story, that if the data were seen in their context it would show something different.

I didn't actually say that. Its true, but what I said was your argument that if they stop doing the things they do well and keep doing the things they do poorly then X will result, is a silly argument to have.

It is up to YOU to support that claim. You can't just throw it out there and expect others to do your research for you.

I don't feel research is necessary. I said
I just don't agree that you can take cumulative statistics, distregard situation, then create a sitaution and apply the cumulative stats to them.
I think that comment stands alone.
You said you don;t disagree then gave me an assignment.
I don't need to complete an assignment to know what you disputed (or didnt because you said you dont disagree) isn't really disputable.

You want to argue that point? Fine. Prove it.

Don't want to argue. Made a point, and it is proven to my satisfaction.

I don't give a crap if you find much value it in.

Then why are you spending so much energy arguing about it?
If you want to give me an assignment the fact that I have no interest in it certainly should matter to you.

If you're off in your own little world convinced that the data doesn't really show what it certainly appears to show, that's ok with me.

What does it certainly appear to show? That if they continue to do what they do poorly equally as poorly and then get bad at what they do well, they will be worse?
I agree. I also agree that all of the other 31 teams will allow a lot of points if they stink in the red zone and don't force turnovers. Luckily for us, we are good in those areas.
I would also say that the teams that don't allow many yards and aren't good in the red zone and don't force turnovers will allow more points if they start allowing a lot of yards.
I'm not sure why any of that is relevant to anything.

But you entered the discussion challenging a point I was making.

And I still do. "If you start to stink at what you are good at and continue to stink at what you are bad at" is a foolish set of parameters for a discussion.
Do you disagree with THAT? Because THAT is what you created this missive over.

t appears that *you* are trying to convince *me* of something.

Only what i just typed abpve. I didnt really think it required convincing.

I'm willing to be convinced. But I'm not going to do your job for you. You want me to be convinced? Convince me.

I don't much care. I am fine that my opinion on this one is sound.

Give me data. If you *aren't* trying to convince me, then why on earth did you bother jumping into the discussion to offer that "correction"?

Because as I said above, your point was silly. If that isn't clear to you by now, have a nice day, maybe we can find something else to discuss at a later date.

 

[Ring 6]

Most of the unaccounted-for variables you bring up would apply equally to per-drive and per-game stats, as well as to yardage and point stats. If you're looking for metrics that allow for these factors, I'd point you to FootballOutsider's DVOA et al., or to the stats kept at AdvancedNFLStats.com, particularly Win Probability based stats, as this metric is entirely based upon context-sensitive calculations of the net change in the likelihood of winning, as established by the results of the last 20 years of NFL football.

As for the per-drive v. per-game question:

Assuming a static level of offensive production, a style of defense that results in fewer possessions and might give up fewer points, but since it also results in fewer possessions for its offense, it needs to give up fewer points to result in the same number of wins.

Essentially, changing the number of possessions in the game creates an 'exchange rate' for points -- a touchdown in a 10 possession game is worth a lot more than a touchdown in a 14 possession game.

Per-game stats don't account for this. Per-drive stats avoid the issue by not grouping possessions by game.

Both stats are equally affected by things like garbage-time, level of competition, etc. Per-drive stats just have one fewer distorting variable.

I can see merit that per drive stats may make a better apples to apples comparison. But I don't accept that a less flawed method or even the least flawed method equals a good method.
Any analysis based upon cumulative statistics, in different situations vs different opponents under different schemes and philosophies that get applied as an 'overall' assessment is really just a poor and flawed method.
The point of playing a football game is to win. There are many different strategies to win, to protect a lead, to get a lead, to avoid collapse, etc.
Every team built all of their cumulative statistics, not only under a different philosophy and scheme, but also under different conditions.
The most important variable in a football team is how it plays in clutch situations that decide games. That just can't be measured in cumulative statistics.
Too often people try to use statistics to predict what will happen when it has already happened. For example in this thread people are trying to use cumulative statistic (a decent portion earned when the game was out of reach) that do not include points allowed to predict how many points the team will allow in the future. Thats silly IMO because you have the exact results those 'supporting stats' actually added up to.
I could care less if the Patriots are the worst defense in the history of the NFL when leading by 28 points with 8 minutes left, because we will never lose a game that way. I recognize that a better defense could play soft and still make plays, but I just don't really have much worry that they are good enough to make plays in that scenario when the game is over. Cumulative stats applied to generic situations do not tell you much.

 

[ivanvamp]

Andy, I guess I just don't understand why you bothered to chime in. You jumped in to correct my thinking about the Pats' defensive performance. Obviously you wanted me to change what I thought about it. I then said I'm willing to be convinced, but you need to give me actual data to think about. You then say that you don't need to because you have enough eye-test evidence to satisfy you.

But....if this was all about satisfying you and not convincing me, why even enter the discussion? It really makes no sense. I mean, I guess if you like hearing yourself talk (or reading your own writing, as the case may be), and that somehow makes you feel good, be my guest. But if your purpose was to engage in meaningful conversation and if you really meant to attempt to change my thinking on something you think I'm in error about, it's an odd way to go about it by not, you know, actually producing evidence and instead just claiming that you yourself are satisfied.

But maybe I'm weird that way.

have a nice day, maybe we can find something else to discuss at a later date.

And to you too. I mean that sincerely.